Upgraded Demos - Touchless Phone Locking/Unlocking

At this point, our first stab demo shows that simply anyone can just pick it up and use it with no trouble at all. Despite the adaptability and compatibility of the device, our initial demo does not really give you any indication of the stability of the algorithm.  Also, having 3 different color screens was unnecessary and a little distracting in fact.  We went back to the drawing board and cut the number of screens down to just black and green.  Black for off.  Green for on. In Demo V2.0, we focused on a few important things.  First, we leave a clock running in the background to show that our demos aren't just a product of slick video editing.  Second, we were trying to demonstrate the stability of the algorithm against accidental locking or unlocking.

 

For Part 2 of the video, we wanted to show that the algorithm was stable when jostled around in a bag.

OK, on the surface, we thought that all of this was pretty cool.  Then reality hit...  When the phone is in the pocket or bag, no one could actually see what the screen was doing!  On, off, no one was the wiser.  We received some friendly advice that we are very grateful for:  Why not just transmit the screen changes to another device via WiFi, Bluetooth, or something?  As long as whatever the phone screen is doing is displayed elsewhere, we could closeout our video demo. To solve this problem, we made the smartphone screen changes into a chat message that we could send through the internet and it would pop right up on a laptop screen.  Same screen colors, green for on, black for off.  This time, we added headphones into the mix.  Even with the added obstruction of the headphone wire, the algorithm works just fine, even with the headphone jack being on the upper part of the phone.  By pure accident, the phone was actually dropped during the demo (perhaps due distracted by bike-pushing person in background).  The locking mechanism was not triggered, and everything worked just fine after that.  Next, we put the phone in the bag and show that our algorithm resists unlocking the phone even when someone is walking around.  So, here it is:

Added advantages.  For a technology like this to be functional, it has to be "always-on."  This means that the accelerometer has to continue to feed data to a processor and work to minimize battery drain.  As we put into our previous posts, after much trial and error, it turned out that an algorithm with minimal complexity and computational load was the best solution.  In addition, our algorithm also uses very little data in order to be functional, reducing the need for memory stores to be active alongside the processor. Where to from here?  Because of the "leanness" of our invention, it could easily be retrofitted to existing devices.  Our only stumbling block is that it would require access to the operating system via a firmware upgrade.